Author |
Message |
Jack Conrad (Jackconrad)
Rating: N/A Votes: 0 (Vote!) | Posted on Saturday, July 23, 2005 - 7:33 pm: | |
I am finishing a new instrument panel using LEDs for indicator and warning lights. Since these LEDs operate on about 2-2.2 volts and about 20 milliamp, I am using a 680 ohm resistor on the 12 volt circuits and a 1.5K omh resistor on the 24 volt circuits. Thay are all woking fine. I would like to cover the bare LED leads with heat shrink tubing. My question is "Can I cover the resistor or will it get to hot?" Thanks in advance, Jack |
Marc Bourget
Rating: N/A Votes: 0 (Vote!) | Posted on Saturday, July 23, 2005 - 8:40 pm: | |
Not only do not cover the resistor, but it's frequently recommended that you attach it to a heat radiator to aid in its heat rejection. The explained purpose is to prolong the life of the resistor. |
Chuck Newman (Chuck_newman)
Rating: N/A Votes: 0 (Vote!) | Posted on Saturday, July 23, 2005 - 9:47 pm: | |
Jack, At 12 vdc the LED is drawing 17.6 MA which results in .21 watts heating in the resistor. At 24 vdc the LED is drawing 16 MA which results in .384 watts heating in the resistor. At 12 vdc, you can get by with a "quarter watt" resitor of 680 ohms. At 24 vdc, you will need a "half watt" resistor. If you use half watt resistors on both voltages, you can shrink wrap the resitors, but it's not industry standard. Let it radiate. I recommend wraping the leads as you have indicated, but leave all the components open to air for the reason Marc stated. Chuck Newman Oroville, CA |
Craig (Ceieio)
Rating: N/A Votes: 0 (Vote!) | Posted on Saturday, July 23, 2005 - 10:37 pm: | |
Jack - you may want to bump up your resistor value a bit to protect your LED from voltage surges. Remember that charging voltages are higher, say 13.8 on the 12v side and 28+v on the 24v side. If (invoke ward-of-protection here) you should lose your regulator you do not want to have to replace the LED's. I have seen automotive regulators fail and allow 18 volts through, so if you were to use that value as you saftey factor, a 1K resistor will keep you below 20 mils of current and you won't pop your LED's. Most silicone diodes have a voltage drop of 0.7v; can you verify the 2 volt spec on yours?. Running an overvoltage will create a reliability issue for you. (It could be that you have three in series and then you are closer to the typical mark). Best Regards, Craig - MC7 Oregon |
John Jewett (Jayjay)
Rating: N/A Votes: 0 (Vote!) | Posted on Sunday, July 24, 2005 - 12:56 am: | |
Why not use a VR7902 voltage regulator and not have to worry about over-voltages up to 36 volts. Admittedly at $.90 cents each, as opposed to $.15 cents for a resistor, they are really a bit pricey, but if you put Paula on an extra paper route or maybe picking up soda cans, you could probably afford a couple. See you on the way back South, in the fall...JJ |
Abajaba
Rating: N/A Votes: 0 (Vote!) | Posted on Sunday, July 24, 2005 - 1:52 am: | |
Probably best to stick with resistors. KISS principle in action here. Normally the body of the resistor is non conductive so there would be no real reason to wrap the sucker with heat shrink. Also, the dissipation ratings are in free air without any insulation around the device. Someone suggested putting the resistor on a heat sink or something. There are some thermal conductive epoxies out there. This would anchor the resistor against vehicle vibration and allow better heat transfer to the heat sink. Might not be a bad idea to look up the thermal conductive epoxy. And if there is some concern about going over the voltage rating of the LED, put a zener in parallel with the LED in the reverse direction so that it doesn't conduct unless the applied voltage on the LED goes too high. The LED still works but the LED does not over voltage. |
Jack Conrad (Jackconrad)
Rating: N/A Votes: 0 (Vote!) | Posted on Sunday, July 24, 2005 - 8:00 am: | |
I used to voltage and amperage specs for each type LED from the catalog. I used 14.8 and 29 volts as the power supply voltage. I am using several type of LEDs so I actually have several different size resisitors in use. I am using red, blue, green, red flashing and reg/green LEDs for different applications. Engine warning (hot engine, low oil, and low air) are red flashing. Jake brake and defroster are green on low and change to red on high. High beam indicator is blue, turn signal indicators are green and all "switch on" LEDs are red. Jack |
Stephen Fessenden (Sffess)
Rating: N/A Votes: 0 (Vote!) | Posted on Sunday, July 24, 2005 - 1:26 pm: | |
Jack, you could run more of the LEDs in series and use corresponding lower value resistores. On 12 volts, you should be able to run 4 LEDs on one string with a resistance calculated for the remaining 5 volts as 25 Ohms and the resistor wattage at 5 Volts X .020 Amps = 0.1 watt. for 24 volts, 11 LEDs with a 25 Ohm 1/4 watt resistor. I would consider lowering the current draw of the LEDs with a higher resistance, since even at 10 ma (with 50 Ohms and the same strings of LEDs) most of LEDs with a 2 volt voltage drop will produce plenty of light. If there can be two or more LEDs on an instrument, use a separate series circuit for each one. But it looks like you are already done. You can also use the liquid electrical tape available at home depot. It is OK to cover the resistor leads, but not he resistor. Keep the resitor leads long and bent at near 90 degrees to stand the resistor off in open air an inch or so away from the panel and instruments. You produce a lot more heat and use much more current using one resistor for each LED. |
Gary Stadler (Boogiethecat)
Rating: N/A Votes: 0 (Vote!) | Posted on Sunday, July 24, 2005 - 1:45 pm: | |
Jack, as electronic components go, LED's and resistors are some of the most robust components out there, and quite hard to kill. Many of the suggestions above are ok, a few are overkill, and a few are not good ideas. The bottom line, if you use the resistance values you are using, and use 1/2 watt resistors as Chuck suggests, you can safely shrink wrap them. Here's what will happen to your LEDs and resistor if: You get overvoltage spikes; nothing You have a failed alternator regulator; nothing (except maybe a lot of other problems not related to the LEDs) You use a (VR7902?? what's that??) solid state voltage regulator in the circuit; lots of extra parts and trouble for nothing The shrink wrapped resistor gets too hot; nothing, because it won't if you use a 1/2 watt size The LED gets over 20ma for whatever reason; nothing (Led's don't die even with 50-100ma going thru them for short times) You're on the right track and almost anything you do here will work fine as you are now seeing. And heck, if one actually fails, which is extremely unlikely with the components you are now using, is it any worse than a burned out lamp? Not really.... |
Craig (Ceieio)
Rating: N/A Votes: 0 (Vote!) | Posted on Sunday, July 24, 2005 - 2:37 pm: | |
Jack - sounds like a fun project. Looks like you verified the spec alright so go with what you have and use the 1/2 watt resistors suggested above. You can heat shrink the leads to keep them from shorting, but don't sweat the resistor body. I have to admit that in my days of designing super computers we did pop a few LED's, so I am conservative with them from a design stand point. Time has marched on and LED technology has changed greatly, so it may be time for me to "update my paradyme" to current technology. The last computer design I worked on was in the late 1980's so... (on the other hand, one of the machines from 1984 is still running with no burned out LED's at the Smithonian so maybe there is something to it!) :-) Have fun and be sure to post pics when you are done. Craig - MC7 Oregon |
rhbelter
Rating: N/A Votes: 0 (Vote!) | Posted on Monday, July 25, 2005 - 11:45 pm: | |
Ho, Jack, I have LED "idiot lights" all over the place. A problem which I have had is some "phantom" lighting of them due to leakage paths. I ended up "strapping" them to 12vdc, (or ground) with a 720 ohm resistor across the LED to dump the leakage. Works OK, but is a bit more work. Enjoy/s/Bob |
|